KR19990047120A - Feature Layer Management System of Feature-based Geographic Information System and Its Method - Google Patents

Abstract

TECHNICAL FIELD The present invention relates to GIS software running on the World Wide Web, and more particularly, to a feature hierarchy management method and system capable of efficiently handling features and automatically determining the types of features that are worth transmitting. .

Assign a level representative value (R-Value) to all features, cluster the level representative value (R-Value) into several groups, and assign a new level representative value (R-Value) to each clustered level. By constructing a hierarchical table (FHT), the levels are assigned according to the spatial hierarchy management that organizes the hierarchy according to the average of the actual size of the geographic elements constituting the feature, and the relative relationship and importance between the features, and then each level. A feature hierarchy management system of a feature-based geographic information system that assigns a level representative value (R-Value) of a proper distribution to a semantic hierarchy that forms a hierarchy according to the relative interrelationship and meaning of features, and It is characterized by providing a method.

Description

Feature Layer Management System of Feature-based Geographic Information System and Its Method

The present invention relates to feature-based GIS software running on the World Wide Web, and more particularly, a GIS capable of handling a large amount of vector data and various kinds of features running on the World Wide Web (WWW). A feature hierarchy management method and system for efficiently handling features for software and at the same time automatically determining the types of features worth transferring.

As the use of WWW grows rapidly, GIS software running on WWW is now beginning to be introduced at several WWW sites. Most of the GIS software for WWW introduced up to now allows users to perform simple searches by connecting attributes to digitized vector data of points, lines, and faces.

For example, a list of buttons for selecting some map elements such as a school, a road, a hospital, a fire station, and the like is displayed on the screen. At this time, the "map element" is also called "feature", and the GIS based on such a feature may be called "feature-based GIS". The feature here is similar to an object, and the term "feature based" has a weaker meaning than the term "object oriented". In other words, "feature-based GIS" can be said to be a predecessor of "object-oriented GIS."

In feature-based GIS, features literally play a central role, and all GIS operations such as spatial analysis and search are performed in feature units. GIS, by its nature, handles large amounts of data. This means that you have to deal with many kinds of features (about 300 or more). Therefore, there is a need for a method that can handle such large-capacity features appropriately and efficiently.

The GIS software for the WWW introduced to date only handled less than about 10 features. Therefore, in this case, the selection buttons of the features may be arranged to allow the user to select a desired feature. But if you have to deal with hundreds of features, you can't use this approach and you need to provide a way for users to understand features and manage them efficiently.

One of the biggest problems with GIS for WWW is that it takes a lot of time to receive large amounts of vector data from the server via WWW. Therefore, a method for controlling the amount of transmission of a large amount of vector data should be prepared.

Some of the GIS software for the WWW introduced so far covers only a few (about 10) specified features, so that traditional stand-alone commercial GIS software for desktops (especially object-oriented GIS software, for example, "Small World"). As in the case of "), there is no feature-based GIS software for the WWW that can handle a variety of features (hundreds).

As described above, for feature-based GIS software that can handle various kinds of features and large amounts of vector data, it is necessary to provide a method for users to handle numerous types of features in a clear and efficient manner. At the same time, the amount of data transmitted through the WWW should be properly controlled.

Accordingly, the present invention is to provide a method and system for managing feature layers. The feature hierarchy management method according to the present invention manages a feature by forming a spatial hierarchy and a semantic hierarchy based on the relationship between the graphic elements constituting the feature and the relationship between the features. By managing features hierarchically, it is possible to control the amount of data to be transmitted, and provide users with visual information about features to enable more efficient and effective feature management.

The present invention makes it easy to manage features by dividing a large number of features into several hierarchies, thereby facilitating user management, and informing a user of candidates for features to be displayed in accordance with an enlargement and reduction ratio of a screen. By preventing the transmission of unnecessary features.

The present invention is a view-by-scale concept used in some existing stand-alone GIS software, that is, when the screen magnification ratio is small, the screen does not display small size vector data, The display method when enlarged is applied to automatically control the management and transfer of features in the feature-based GIS software for WWW.

1 is a schematic view showing a feature hierarchy management method according to the present invention.

2 is a structure diagram of a feature hierarchy table (FHT) according to the present invention.

3 is a flowchart illustrating a method of reallocating the visible-flag of the FHT when the screen is enlarged / reduced according to the present invention.

<Description of the code | symbol about the principal part of drawing>

10: database 11: database management system

12: Web server 13: Java class file

110: feature hierarchy control unit 121-123: feature hierarchy table

130: layer module 140: zone module

140: project module

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The core of the present invention is related to the concept of managing various types of features divided into layers and a method of automatically configuring the layers.

Two methods are presented to automatically construct the hierarchy. One method is to construct a hierarchy according to the size of a feature. This is called spatial hierarchy, and the other is to construct a hierarchy according to the meaning of the feature. This is called semantic hierarchy.

1 is a block diagram of a feature hierarchy management system according to the present invention.

A database (DB) 10 in which all attribute data and graphic data related to the feature are stored and constructed in the server, a database management system (DBMS) 11 for managing the database 10, Feature layer control unit (FHCU) for determining and adjusting the type of data by the visible-flag (V-Flag) value of the feature layer table (FHT) when the geographic element data stored in the database 10 is imported into a web browser. 110, a number of feature hierarchy tables (FHTs) 121, 122, and 123 that store information such as levels, feature names, level representatives, visible-flags, include-flags, and the like. Project module 150 that processes hierarchical information about all the features to be added, and inserts, deletes, and modifies the feature into a vertically cut region of any desired region in the feature set of the project module 150. Above Zone module 140, a layer module 130 for processing a topological relationship or spatial analysis for a single feature among several kinds of features in the zone module, and the project module 150. The classes of the zone modules 140 and the layer modules 130, the classes of the feature hierarchy table 121-123 and the feature hierarchy control unit 110, are stored in the Java class file 13, and the executable files It is composed of a web server 12 provided in the respective parts in the form of a class.

In the feature layer management system of the present invention thus configured, the executable files of the system are provided in the web server 12 in the form of a class. In the Web server 12, the classes and feature hierarchy table (FHT) of the project module 150, the zone module 140, and the layer module 130, which can be referred to as the basic configuration modules of the system. 123 and class of feature layer control unit (FHCU) 110. The data transfer from the database (DB) 10 through the database management system (DBMS) 11 to the client's web browser is controlled by the feature layer control unit (FHCU) 110, in which case each project module ( 150, the FHTs 123-121 of the zone module 140 and the layer module 130 are referred to.

The project module 150 corresponds to a collection of all features managed by the system. Zones in which any desired area is vertically cut out from the feature set of the project module 150 correspond to zones. Inserts, deletions, and modifications to features occur in the zone. There are many kinds of features in a zone, and it is necessary to perform topological establishment or spatial analysis on a single feature. In this case, one feature may be viewed and processed as one layer, and inter-layer calculation such as overlay may be performed using a plurality of layers composed of each feature.

The feature hierarchy table (FHT) 121-123 is composed of fields of level, feature-name, level representative value (R-Value), visible-flag (V-Flag), and containing-flag (C-Flag). Table. Levels refer to levels on a feature hierarchy, and the number of levels may be predetermined or adaptively determined according to the distribution of data. Multiple feature-names means that multiple features can be assigned to a level. The level representative value (R-Value) represents a representative value (Representative value) of the level, which is determined in consideration of the spatial size relationship of the feature. How many levels a whole feature set is classified is determined by looking at the distribution of the level representative values of the feature set. Visibility-flag (V-Flag) refers to the visibility flag (Visibility flag) to represent the visible information of the features of the level. In addition, the containment-flag (C-Flag) means a containment flag (Containment flag) and indicates whether the feature is within the current screen range. This is shown in [Table 1] shown below.

Level Feature-Names Level representative value (R-Value) Visible-flag Include-flag (C-Flag)

The feature layer control unit (FHCU) controls the transmission of geographic element data from the DB to the web browser, and whether or not to transmit the data is determined by the V-Flag value of the FHT.

On the other hand, the features are ordered by a set of rules, which in turn determine the loading and display of data from the server. The present invention proposes two methods of constructing features.

The first is to construct a hierarchy based on the actual average size of the geographic elements that make up the feature, which is called a “spatial hierarchy” (300), which is the basis of the feature hierarchy because it is a classification according to the spatial size of the feature. can do. The second is called "Semantic hierarchy 400" and constructs the hierarchy according to the relative interrelation and meaning of the features. Unlike the first method, this method is based on semantic and relative size comparison between features regardless of the actual size of the geographic elements that make up the feature. It also takes into account the semantic significance of the feature, regardless of its size.

2 is a structure diagram of a feature hierarchy table according to the present invention.

The method for constructing the FHT 500, in the case of the spatial layer 300, first calculates the R-Value for all features (310). The FHT 500 is configured by grouping the R-Value into several groups and then assigning a new R-Value to each level (320). Meanwhile, in the case of semantic hierarchy 400, a level is allocated according to the relative relationship and importance between features, and so on (410). Next, an appropriate distribution of R-Values is assigned to each level (420).

In addition, the method of importing the feature to the web browser of the client should reassign the V-Flag of the FHT when the screen is enlarged / reduced. This is illustrated in FIG. 3.

When the zoom ratio of the screen is changed, the V-Flag of the FHT must be updated first. When the ratio of the screen zoom is changed (S1), the visible flag V-Flag is cleared (S2), and the lowest level feature among the unselected features is selected (S3). In this way, the lower level of the feature layer is selected and the ratio of its R-Value to the current screen size is compared with the threshold TH (S4). If the value is larger than the given threshold, the level V-Flag is set to on (S5). For example, if the ratio of the current screen size and the R-Value is less than or equal to 10, there may be a method of setting the V-Flag on. If V-Flag is off at a certain level, all levels below it are set off (S6). In this case, however, the FHT should be aligned according to the level. In this way, if all V-Flags of the FHT are set for all the features, only the features of the level where the V-Flag is on are loaded and displayed (S7).

Spatial Hierarchy Management

You need to set the level representative value (R-Value) for each feature. In order to determine the representative value of each feature, an intra feature relation analysis on the size of the geographic elements of the feature must be preceded. There are three types of geographic elements: nodes, chains, and polygons. It is convenient to use the Minimum Bounding Rectangle (MBR) to indicate the size of the three elements. In addition, there may be various ways to indicate the size of the MBR. In other words,

1. The larger of the width and height of MBR (max (width, height))

2. Sum of width and height of MBR (width + height)

Diagonal length of MBR (sqrt (width2 + height2))

4.Area of MBR (width * height)

Etc. is possible. Select one of these, average it, and determine it as the R-Value of that feature.

The method of determining the average value of the size of the geographic elements constituting the feature as the representative value of the feature is based on the assumption that the intra feature variation of the feature with respect to the size between the geographic elements constituting the same feature is not large. If the variance in the size between the geographic elements that make up a single feature is large, there will be a lot of methods to determine whether to load and display by feature by assigning a representative value to the feature.

There are many kinds of features that can appear on the drawings. U.S. In the feature code list provided by Geological Survey, more than 350 features appear. In the "DIGEST's Feature Attribute Coding Catalog" you will find a wide variety of features. There are some problems with assigning and managing unique levels for each of these features. The first is that the number of records in the FHT becomes too large. The second is that the number of comparison operations between the R-Value and the screen size increases when determining whether to load data.

Therefore, by grouping features having similar R-values at the same level, it is possible to reduce the number of unnecessary comparison operations and to maintain the number of levels of the entire hierarchy, thereby facilitating the management of the FHT. The number of levels may be predetermined or adaptively determined by analyzing the spatial distribution of the feature.

The fixed number of levels method, which predetermines the number of levels, is a method of assigning features to several predetermined levels. The number of levels will have to be determined appropriately considering the number of features. Supervisor clustering methods such as the k-means algorithm can be used to cluster R-Values into k levels.

A variable number of level method of analyzing the spatial distribution of features and adaptively assigning levels is to dynamically determine the number of levels according to the distribution of R-Values. That is, clustering features with similar R-Values at the same level. This has the advantage that the FHT can be automatically configured without prior knowledge about the distribution of features and R-Value. Un-supervised clustering methods such as agglomerative algorithms can be used. If the result of clustering is that all features are assigned to only a few levels, a method of reassigning them to an appropriate number of levels is required.

Semantic Hierarchy Management

Regardless of the spatial size of a feature, a hierarchy may be constructed by semantic information of the feature. For example, the city boundary feature is smaller than the state boundary feature and the state boundary feature is always smaller than the state boundary feature. This semantic hierarchy will list all possible features. Ambiguous features to determine hierarchy are assigned at the same level. FHT can be configured by giving an appropriate R-Value to each level of the hierarchical structure.

One noteworthy of semantic hierarchy management is the assignment of point element features that have significant meaning to higher levels. In terms of size, we assign features that belong to lower levels to higher levels so that they load and display even at lower zoom ratios. For example, important public institutions and emergency facilities, although only point information having location information, are assigned to a high level and displayed to the user first.

Therefore, semantic hierarchy management is a method of constructing hierarchy considering not only the relative information of the feature size but also the importance of the feature.

As described in detail above, the present invention manages a feature by forming a spatial layer and a semantic layer based on the relationship between the graphic elements constituting the feature and the relationship between the features, thereby controlling the amount of data to be transmitted and allowing the user to Providing visualization information about features has the effect of enabling more efficient and effective feature management.

Claims (11)

A project module 150 for processing collection information on all features managed by the system, A zone module 140 for inserting, deleting, and modifying a feature for an area in which a desired region is cut out of the feature set of the project module 150; A layer module 130 for processing topological relations or spatial analysis of a single feature among the various types of features in the zone module, A feature hierarchy table (FHT) 121, 122, and 123 that stores information such as level, feature name, representative value of the level, visible-flag, and include-flag for the feature hierarchy managed by each module; In order to create the feature hierarchy table, spatial hierarchy management for constructing and managing hierarchical groups that are clustered according to the average size of geographic elements, and semantic hierarchy management for constructing and managing hierarchies according to the relative interrelation and meaning of features The feature layer control unit (FHCU) determines and adjusts the type of data based on the V-Flag value of the feature layer table (FHT) when importing geographic data stored in the database into a web browser. ) 110, A database (DB) 10 which receives an executable file of each part from a web server 12 in the form of a class, and which each part is built on a web browser of a client and stores all attribute data and graphic data related to a feature. Feature hierarchical management system of a feature-based geographic information system, characterized in that the hierarchical management by importing features from the. The method of claim 1, wherein the feature layer table (FHT), A level field means a level on a feature hierarchy and the number of levels is predetermined or adaptively determined according to the distribution of data; The name of a feature, and a feature-name field where multiple features can be assigned to one level, A level representative value (R-Value) field, which is representative of the level and is determined in consideration of the spatial size relation of the feature, and is used to determine how many levels the entire feature set is to be classified; A V-Flag field indicating visual information of features at the corresponding level, A feature hierarchy management system of a feature-based geographic information system, characterized in that the table consists of a C-Flag field indicating whether or not the feature is within the current screen range. The method of claim 1, wherein the spatial hierarchy management, Assign a level representative value (R-Value) to all features, After the level representative value (R-Value) is clustered into several groups, a new level representative value (R-Value) is assigned to each clustered level to construct a feature hierarchy table (FHT). A feature hierarchy management system of a feature-based geographic information system, characterized in that it is managed in a spatial hierarchy constituting a hierarchy according to the average of the actual size of the geographic elements constituting the feature. The method of claim 1, wherein the semantic hierarchy management Assign levels based on their relative relationship and importance, Next, assign a level representative value (R-Value) of the appropriate distribution to each level. A feature hierarchy management system of a feature-based geographic information system, characterized in that it is managed by a semantic hierarchy constituting a hierarchy according to the relative interrelation and meaning of features. Assign a level representative value (R-Value) to all features, After the level representative value (R-Value) is clustered into several groups, the feature hierarchy table (FHT) is constructed by assigning a new level representative value (R-Value) to each clustered level. A method for managing a feature hierarchy of a feature-based geographic information system, characterized in that it is managed as a spatial hierarchy constituting a hierarchy according to an average of the actual size of geographic elements constituting the feature. The method of claim 5, wherein the method of determining a representative value (R-Value) of the feature comprises: Determined by intra feature relation analysis of the size of the geographic elements of the feature, Consider three types of geographic elements: node, chain, polygon. The minimum bounding rectangle (MBR) is used to represent the size of the three elements. As a method of indicating the size of the MBR, a). Larger value of the width and height of MBR (max (width, height)) b). Sum of width and height of MBR (width + height) c). Diagonal length of MBR (sqrt (width2 + height2)) d). Area of MBR (width * height) A method of managing a feature hierarchy of a feature-based geographic information system, characterized in that one of the following methods is selected and then averaged to determine the R-value of the feature. The method of claim 6, wherein the method of determining the average value of the size of the geographic elements constituting the feature as the representative value of the feature comprises: A method for managing a feature hierarchy of a feature-based geographic information system, characterized in that the intra feature variation of a feature is not large with respect to the size of geographic elements constituting the same feature. The method of claim 5, wherein the level representative value is clustered. Features with similar sized R-Values are grouped together at the same level, The fixed number of level method is used to assign the R-Value to k levels predetermined by the user by the number of features using the supervisor clustering method. Feature hierarchy management method of feature-based geographic information system. The method of claim 5, wherein the level representative value is clustered. Cluster features with similar R-Values at the same level, A method of managing a feature hierarchy of a feature-based geographic information system, comprising using a variable number of level method for adaptively assigning levels by analyzing spatial distribution of features. Assign levels based on their relative relationship and importance, Next, assign a level representative value (R-Value) of the appropriate distribution to each level. A method for managing a feature hierarchy of a feature-based geographic information system, characterized in that it is managed by a semantic hierarchy constituting a hierarchy according to the relative interrelation and meaning of the feature. The method of claim 10, wherein the semantic layer management, A feature hierarchy management method of a feature-based geographic information system, characterized by assigning hierarchical features to the same level and assigning important point element features to higher levels.